Established models of rapid-eye movement sleep (REMS) advocate the involvement of brainstem cholinergic and monoaminergic neurons, although lesions of these nuclei do not markedly alter REMS suggesting the involvement of other neural circuitry in this behavior. Our laboratory has an alternative model of REMS based upon the following findings: lesions of the ventrolateral periaqueductal grey (vIPAG) and lateral pontine tegmentum (LPT) increase REMS, while the sublaterodorsal nucleus (SLD) and precoeruleus (PC), which send and receive extensive GABAergic inputs to and from the vIPAG and LPT, contain REM- active neurons and when lesioned, diminish REMS (Lu et al. 2006). These observations suggest the existence of a flip-flop circuit whereby REM-off neurons in the vIPAG-LPT and REM-on neurons in the PC- SLD regulate REMS through mutual inhibition. Thus, our hypothesis predicts that GABAergic neurons in the SLD act as REM-on neurons by inhibiting the REMS-off side of the switch. To assess this model, I will record REMS in conditional knockout mice that receive nucleus-specific ablation of GABA release in the SLD through inactivation of localized expression of vesicular GABA transporters (VGAT) necessary for packaging GABA for synaptic release. Specifically, Cre will be injected with a viral vector (AAV) into these brain regions to remove loxP-flanked segments of DNA coding for localized VGAT. The loss of GABA release from SLD neurons should reduce REMS and result in more transitions in and out of REM sleep. Lu et al. (2006) also shows that ventral SLD (vSLD) lesions reduce muscle atonia during REM sleep. Since direct glutamatergic spinal projections from the vSLD are implicated in this phenotype, I propose to assess REM sleep in conditional knockout mice in which glutamate release in the vSLD is decreased through inactivation of localized expression of glutamatergic vesicular transporters (vGLUT2), the vesicular transporter necessary for packaging glutamate for synaptic release in this brain region. I hypothesize that mice which lack glutamate in the vSLD will exhibit reduced muscie atonia during REM sleep. The brain initiates and controls REM sleep. These studies will further our understanding of how the brain accomplishes this work. This understanding is necessary for the development of drugs designed specifically to treat REM sleep disorders. [unreadable] [unreadable] [unreadable]